Proton Radiotherapy for Pediatric Medulloblastoma: Improved Early Ototoxicity

Reviewer: Geoffrey Geiger, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: October 31, 2010

Authors: B. J. Moeller, M. Chintagumpala, J. J. Philip, S. Y. Woo, J. E. Wolff, A. Mahajan
Institution: M.D. Anderson Cancer Center, Houston, TX; Baylor College of Medicine, Houston, TX; University of Louisville School of Medicine, Louisville, KY


  • Medulloblastoma represents approximately 1 in 5 pediatric central nervous system tumors.
  • Males are more frequently affected than females in a 2 to 1 ratio with a median age of 5-6.
  • At the time of diagnosis, patients are divided into standard and high-risk medulloblastoma, based on involvement of the CNS outside of the tumor bed, of which 1/3 are high-risk.
  • Ototoxicity is an easily quantitated and common side effect cranial irradiation and platinum-based chemotherapy, and occurs in approximately half of patients.
    • Ototoxicity can profoundly affect cognitive and social development in this population.
      • Ototoxicity is multifactorial in nature since patients frequently receive trimodality therapy with maximal surgical resection followed by chemotherapy and radiation.
    • A large body of data exists demonstrating that higher radiation doses to the cochlea (particularly above 36 Gy) may result in greater ototoxicity as measured by audiometric outcomes.
  • Patients have been historically treated with photon radiation but more recently, investigations into the use of proton radiotherapy have been performed, taking advantage of the superior dose characteristics in an attempt to reduce integral dose and reduce long-term side effects.
    • One benefit of use of proton radiation to deliver boost radiation to the posterior fossa/ tumor bed is relative sparing of the cochlea.
    • Proton radiotherapy has superior dose characteristics, and plans generated with protons have superior DVH parameters than intensity modulated radiotherapy (IMRT) plans, which may help reduce the development of ototoxicity in pediatric patients undergoing radiation therapy for medulloblastoma.
  • This investigation was performed to prospectively determine whether audiometric outcomes were better at 1 year in patients receiving proton radiotherapy as compared to IMRT.

Materials and Methods

  • This investigation was conducted as a prospective observational study between 2006 and 2009.
  • Inclusion criteria: age <18, resected medulloblastoma, receipt of cisplatin-based chemotherapy, and appropriate audiometric testing.
    • Patients underwent pre-treatment and 1-year post-treatment pure-tone audiometric evaluation to assess ototoxicity.
  • Patients with moderate-to-severe baseline hearing loss were censored from analysis, which left 28 ears in 15 separate analyzable patients.
  • Raw audiometric data and ototoxicity rates at one year (using Pediatric Oncology Group [POG] guidelines) were compared to a matched historic photon (IMRT)-treated cohort.
    • There were 19 patients in the prospective proton cohort and 15 patients in the retrospective IMRT cohort.
  • The main datapoint was audiometric outcome at 1 year following completion of radiation therapy.


  • The two cohorts were well balanced with respect to patient characteristics and distribution of standard and high-risk patients.
  • For the IMRT- and proton-treated cohorts, the predicted mean cochlear radiation doses were 37 Gy and 30 Cobalt Gray Equivalents (CGE), respectively (range, 23-51 vs. 19-43, p <0.01), and the mean cumulative cisplatin doses were 290 and 303 mg/m2, respectively (range, 180-340 vs. 298-330, not significant).
  • Post-radiotherapy hearing sensitivity significantly declined at higher frequency ranges (2-8 kHz) compared to pre-radiotherapy sensitivity (p <0.05).
    • The observed threshold loss was slightly worse in the right ear compared to the left, consistent with prior reports on radiation ototoxicity.
  • Mean post-radiation audiometry thresholds favored proton- vs. IMRT-treated patients at low-to-moderate frequencies (7 vs. 15, 9 vs. 15, 16 vs. 26, and 25 vs. 30 dB at 1, 2, 4, and 6 kHz).
    • There were significant differences in each examined frequency level for patients treated with protons versus patients treated with IMRT in the audible speech range of 0-4.5 kHz.
  • Predicted rates of severe (POG grade 3 or 4) ototoxicity at 1 year were lower for proton vs. IMRT-treated patients (7 vs. 14%). For proton-treated patients, ototoxicity did not correlate with predicted dose to the auditory apparatus.
  • No significant dose response was seen with respect to audiometric outcomes.

Author's Conclusions

  • This study is the first demonstrating that rates of early high-grade treatment-related ototoxicity in children with medulloblastoma are reduced by half with proton compared to photon (IMRT)-based radiotherapy.
  • Preferential preservation of hearing in the audible speech range may eventually translate into improved quality of life and school performance for children treated with proton-based radiotherapy for medulloblastoma.
  • Continued surveillance of this cohort is needed to determine whether these differences in early ototoxicity correspond to late radiation effects on the cochlea.
  • No dose response was observed in this patient population and therefore, further reduction of dose beyond what can be achieved with proton radiotherapy is not likely to be clinically meaningful.
  • Further efforts are needed to minimize radiation-unrelated ototoxic insults for these patients.

Clinical Implications

  • Preservation of hearing function in patients at risk for cognitive deficits from radiation therapy is critically important and may improve patient quality of life.
  • This investigation is the first to demonstrate that treatment-related ototoxicity in children with medulloblastoma is reduced by half with proton compared to photon (IMRT)-based radiotherapy at one year following completion of radiotherapy.
  • There are some limitations of this study:
    • The control group (IMRT cohort) was retrospectively examined and the proton group was evaluated prospectively, which introduces bias into comparisons of the treatment groups as compared to a study randomizing patients to either proton or IMRT therapy.
    • Ototoxicity at 1 year may be too early an endpoint to fully examine late toxicity from radiation, and further follow-up will be crucial to determine the long-term differences between these groups.
    • Although the authors report on the doses of cisplatin between the two groups, there is no data reported on treated radiation volumes within the two cohorts. If radiation volumes differed significantly in the proton and IMRT groups, this may have altered the dose required to be delivered near or affecting the cochlea.
  • Although no dose response to RT was observed in this investigation, it is likely because most of the data points were below the threshold dose of 36 Gy.
  • Ultimately, this study reports on important early data from a prospective cohort of patients treated with proton radiotherapy examining ototoxicity at 1 year as compared to a similarly matched retrospective cohort treated with IMRT. More long term data from this investigation will be important in ascertaining whether the differences between the two groups persist with longer follow-up.